Synthesizing play: exploring the use of artificial intelligence to evaluate game user experience

Abstract

Digital games are a complex interactive medium providing a multitude of different experiences. The field of games user research (GUR) is dedicated to investigating and optimizing user experience in games. Such inquiries are of both commercial and academic importance, enhancing product quality and our understanding of human behaviour. A common GUR methodology is usertesting, where researchers gain insights from human users interacting with products. However, usertesting is expensive in terms of expert labour, time, and resource costs. To address these concerns, we developed PathOS, a free, open-source tool for game testing with AI agents. PathOS simulates player navigation in games using a basic model of human behaviour. We conducted an evaluation of PathOS with developers, finding that it provides valuable predictions of user behaviour in the iterative design process. Ultimately, we aim to give the game development community a useful and versatile augmentation to their testing processes.